Re: [Patents] Re: Software Patent Directive

[Hartmut Pilch <phm@a2e.de>]

> >     http://swpat.ffii.org/papers/eubsa-swpat0202/kinv/
>
> Can a Washing Machine be a "Computer-Implemented Invention"?
> Only Programs are Computer-Implemented.
> According to Art 52 EPC, programs for computers are not inventions in
> the sense of patent law.
>
> Only Programs are Computer-Implemented is not right, you don't implement
> a program.

I think you are not sticking to the language game of patent doctrines but
rather switching over to programmers' colloquial talk about "implementing
this or that feature" by writing out program code to fully express a
preconceived idea (as e.g. expressed in an interface specification).

In this colloquial usage of programmers and engineers, you don't
"implement" an invention at all.   Or, if you think of the ideas described
in a swpat claim as an "invention", you may be "implementing" them by
doing a certain kind of work, such as programming, rather than by using
a certain kind of machine, such as a computer.

But that is evidently not the language game that we are starting from.

You have to think ab bit more abstract here, not stick to narrow
colloquial usage of words.  We are starting from a strange language game,
set out by patent lawyers, and you have to stay within this language game.

A program can be implemented by a television channel.  A program is a
sequence of events, governed by some rules, "implemented" by some
machinery.  A tv program is also a program.  As is a course for learning.
Or a data processing program.  Or some other construction that may be
described in patent claims.

When the BGH (german federal court of justice) said in its ABS decision of
1980

        http://swpat.ffii.org/papers/bgh-abs80/

that there are "technical programs" which are not mere "programs for
data processing equipment", it was saying that the "ABS program" is a set
of instructions for dealing with forces of nature rather than with data,
and that its validity was verified by experiments with forces of nature,
not by abstract calculus on data.  In that sense one can indeed talk of
"technical program".  But such a program is not "computer-implemented".
Only data-processing programs are "computer-implemented".

Perhaps the words "data processing" should be added into the text on the
web page to make it clearer that we are not talking in the context of
a programmer "implementing a feature in a specification".

-- 
Hartmut Pilch, FFII & Eurolinux Alliance              tel. +49-89-18979927
Protecting Innovation against Patent Inflation       http://swpat.ffii.org/
290,000 votes 2000 firms against software patents    http://noepatents.org/